Biased Estimates of Equilibrium Climate Sensitivity and Transient Climate Response Derived From Historical CMIP6 Simulations.

This study assesses the effective climate sensitivity (EffCS) and transient climate response (TCR) derived from global energy budget constraints within historical simulations of eight CMIP6 global climate models (GCMs). These calculations are enabled by use of the Radiative Forcing Model Intercomparison Project (RFMIP) simulations, which permit accurate quantification of the radiative forcing. Long‐term historical energy budget constraints generally underestimate EffCS from CO2 quadrupling and TCR from CO2 ramping, owing to changes in radiative feedbacks and changes in ocean heat uptake efficiency. Atmospheric GCMs forced by observed warming patterns produce lower values of EffCS that are more in line with those inferred from observed historical energy budget changes. The differences in the EffCS estimates from historical energy budget constraints of models and observations are traced to discrepancies between modeled and observed historical surface warming patterns. Plain Language Summary: Here we use climate models and observations to evaluate the extent to which future warming can be inferred from historical climate change. We assess the historical energy budget in 8 climate models by leveraging a recent community effort to quantify radiative forcing within their historical simulations. We find the historical energy budget tends to provide a biased‐low constraint on future warming, represented by two metrics: effective climate sensitivity and transient climate response. Moreover, observations and simulations run with observed surface warming patterns produce even lower values of climate sensitivity. This difference in climate sensitivity estimates can be traced to discrepancies between modeled and observed surface warming patterns. Key Points: Estimates from historical energy budget constraints underestimate equilibrium climate sensitivity and transient climate response in modelsAtmosphere‐only models forced by observed surface warming patterns produce lower estimates of ECS, in line with historical observationsDiscrepancies between modeled and observed historical surface warming patterns account for the differences in climate sensitivity estimates [ABSTRACT FROM AUTHOR]